# A new generation of interaction models

Research Group:
Centre for Condensed Matter and Material Physics
Number of Students:
1
Length of Study in Years:
3-4
Full-time Project:
Yes
Funding:
CSC
Project Description:
---------------------------------------------------
NOTE: This project has a strong overlap with
"Potentials from the electronic density"
---------------------------------------------------

Intermolecular interaction models form the corner stone of most
simulations of the condensed phase. But most of these models are
simple and often very inaccurate. Many simulations are performed
with Lennard-Jones models, but we have known for decades that these
are just too simple and miss out on a lot of detail that is necessary
if we wish to make our simulations more predictive. We can develop
more accurate models based on ab initio data using methods such as
SAPT(DFT) (a symmetry-adapted perturbation theory based on density
functional theory), but these are not easy to make and generally
incur a considerable computational cost in their development.

Recently, my group, in collaboration with groups in Cambridge and
at the University of Wisconsin-Madison in the US, have developed a
new methodology to derive intermolecular potentials from a relatively
limited amount of ab initio data (molecular properties, some dimer
interaction energies using SAPT(DFT)) and almost no fitting. These
potentials are not only very accurate, but are also very easy to
create. However, they apply between distinct molecular species only.
This is alright if we are dealing with a gas, liquid or molecular
solid. But as we deal with larger systems we need to develop models
that can be applied to the interactions within a complex. Consider
a large protein: it not only interacts with other proteins and the water
and ions that surround it, but parts of a protein also interact
with other parts of the same protein.

In this project we will extend the methods developed for
molecule..molecule interactions to handle cases like this. We will
need to ensure that we capture the effects of intramolecular
van der Waals, polarization, exchange, electrostatic and
charge-transfer interactions. There is good reason to believe that
our very physical models for intermolecular interactions will work
very well in this more complex scenario. At the end of this project
we will have a generalised interaction model that can be applied in
all cases. This will have tremendous potential for applications.

The project will involve a deep understanding of electronic structure
methods, in particular, the theory of intermolecular interactions
through perturbation theory. It will also require a considerable
amount of programming in Fortran90 and Python, and it is very likely
that simulations will need to be performed using codes like DL_POLY.

Candidates need to be strong, or to have a strong desire to learn
the required theory and programming techniques. These methods will be
coded in the CamCASP program. Candidates will also need to be keen to
implement parallel algorithms to best utilise multi-processor computers.

Relevant publications:
SAPT(DFT):
* A. J. Misquitta, R. Podeszwa, B. Jeziorski, and K. Szalewicz,
Intermolecular potentials based on symmetry-adapted perturbation theory
with dispersion energies from time-dependent density functional
calculations.'', J. Chem. Phys., 123, 214103-14 (2005).
* A. J. Misquitta, B. Jeziorski, and K. Szalewicz, Dispersion
energy from density-functional theory description of monomers'',
Phys. Rev. Lett. 91, 033201-4 (2003).

Potentials:
* A. J. Misquitta, G. W. A. Welch, A. J. Stone and S. L. Price,
A first principles prediction of the crystal structure of C6Br2ClFH2'',
Chem. Phys. Lett. 456, 105-109 (2008).
* T. S. Totton, A. J. Misquitta and M. Kraft A first principles
development of a general anisotropic potential for polycyclic aromatic
hydrocarbons'', J. Comp. Theor. Chem., 6, 683-695 (2010).

Requirements:
These are desirable, but may also be acquired during the course of the project:
* Strong programming background in Fortran 90 and Python
* Experience with Linux
* Strong mathematical background
* Knowledge of electronic structure methods like Hartree-Fock and Density Functional Theory.